32 MISC. PUBLICATION 257. U. S. DEPT. OF AGRICULTURE 



be greater, even though the total transpiration is less, than in the sun 



shoot. It is not surprising, therefore, to find that when shade leaves 

 are placed in the sun they wilt more readily than the sun leaves, in 

 spite of a lower absolute water loss. 



The transpiration in turn influences the amount of the conducting 

 tissue, especially the number and width of the wood elements in the 

 vascular bundles, as pointed out by Kohl {127) and later by other 

 authors. Winkler (£££, pp. 65-79) showed that plants which tran- 

 spire heavily build vessels with wide lumina, and vice versa, and came 

 to the conclusion that between the amount of transpiration and the 

 mass of conducting tissue there was a decided correlation. Also, plants 

 which have to absorb relatively large amounts of water in a compara- 

 tively short time have well-developed vascular tissues, as pointed out 

 by Cannon (34) in the desert plants of Arizona. Corillea, Parkinsonian 

 Prosopis, etc., which take in most of their water dur'ng the one or two 

 rainy periods of the year, have a larger number of vessels per unit 

 cross section of stem than those which obtain their water more uni- 

 formly throughout the growing season. Not only are the vessels 

 greater in number but the individual conducting elements are larger, 

 according to Groom (75), who reports that the American species of 

 pine with narrow tracheitis ('0.024 to 0.039 mm) generally live in drier 

 habitats than those with w der tracheitis (up to 0.05 mm). The width 

 of the wood elements in the various species varies with the moistness 

 of the habitat, although it should be noted, as pointed out by Biisgen 

 and Munch {33, p. 182), that there are many exceptions to this general 

 ride, and the determining climatic and edaphic conditions are fre- 

 quently difficult to analyze exactly. The change from spring to fall 

 wood is associated also with the water supply and supports the above 

 conclusions. As to experimental work on this subject, we have the 

 results of Wieler {235 1. who found that small trees of Robinia and 

 Quercus sessiliflora Salisb. in nutrient solut'ons, where there was an 

 abundance of water and a moist atmosphere, produced wood with 

 abnormally wide conducing elements. 



This relation between the conducting system and the ability of trees 

 to rep" ace water losses has been studied also by Stefanoff and StoickofT 

 {208), who reported that in conifers water is conducted to the leaves 

 about as rapidly as it is lost, permitting only slight water deficits. 

 The same is true of Alnus glutinosa Gaert., Castanea vesca Gaert., 

 Acer dasycarpum Ehrh., A platanoides, and Carpinus betulus. The 

 evergreen broad-leaved trees, however, such as Buxus and Primus 

 laurocerasus L., and such trees as Fraxinus oxycarpa Willd., F. excel- 

 sior, Morns alba L.. Gleditsia iriacanthos, Robinia pseudoacacia L., 

 Jv.olans regia L., and Acer pseudoplatanus. show heavy water deficits 

 and cannot conduct water rapidly enough to replace transpiration 

 losses even when an abundant supply is available. 



The characteristic structures of sun leaves are consequently thought 

 by many workers to be produced by drought rather than by light. 

 To test this hypothesis experimentally Soding {197) studied the effect 

 of physiological drought on small potted trees of Tilia cordata Mill., 

 Fraxinus excelsior, and Acer platanoides by adding XaCl to the soil. 

 The trees thus treated had smaller leaves and shorter shoots than the 

 controls and showed all the characteristics of sun leaves and shoots, 

 leading to the conclusion that such modifications are caused by water 

 deficits in the tissues at the time of shoot formation. 



